The Use of Intelligent Multi-Agent Systems Based on Ontologies in the Management of Transport Enterprises

EasyChair Preprint 7648

Abstract

The article discusses the principles of creating intelligent systems for situational management of transport enterprises in real time. A multi-agent approach to the construction of such systems associated with the use of ontological models is proposed. Situational management includes reaction to events, distribution and planning of resources, optimization of the solution, coordination with users, monitoring and control of the implementation of the constructed plan. An increase in management efficiency is associated with the achievement of consensus in the processes of joint decision-making by persons in a problem situation and participating in its settlement. In this context, in order to speed up and improve the decision-making procedure based on consensus, it is proposed to build multi-agent models of situations that provide support for consensus processes. The stages of work that need to be carried out on the way to creating consensus support tools are described. The article presents the experience of semantic modeling of the subject area (transport sphere) based on ontology. As a prerequisite for choosing the basic elements of ontological specifications, the cognitive ability of modeling subjects to distinguish between objects and detect relationships between objects is postulated. The general scheme for the use of ontological models is analyzed, its organic orientation towards the integration of heterogeneous knowledge is stated, and the mechanisms for managing models necessary for this are outlined.

Keyphrases: Actor, Agent, Ontology, decision making, intelligent system, multi-agent technologies

@booklet{EasyChair:7648,
  author    = {Olha Tkachenko and Kostiantyn Tkachenko and Oleksandr Tkachenko},
  title     = {The Use of Intelligent Multi-Agent Systems Based on Ontologies in the Management of Transport Enterprises},
  howpublished = {EasyChair Preprint 7648},
  year      = {EasyChair, 2022}}